- Volume 141, Issue 2, 1995

In the framework of the international collaborative project aiming to sequence the whole Bacillus subtilis chromosome, we have created a relational database for managing and analysing information associated with the molecular genetics of this bacterium: Subtilist. It allows recovery of non-redundant DNA sequences of the B. subtilis genome, as well as related information, i.e. genes, proteins, etc. A logical structure has been designed with appropriate links between the different objects, and a set of procedures has been implemented for data updating and management. The database is organized around a core constituted by all known contigs of B. subtilis, i.e. sets of nonredundant sequences created from original entries in the EMBL data library. A user-friendly interface has been developed to make the database easy to consult. Sequence analysis tools have been integrated into the database, such as a program for rapid similarity searching of protein data banks, and a powerful DNA pattern searching program. Thanks to the consistency of Subtilist, we have performed a codon usage analysis by Factorial Correspondence Analysis, and a study of the distribution of the isoelectric points of known proteins of B. subtilis. The Subti List database is available through anonymous ftp (address ‘ftp.pasteur.fr’ or IP number 157.99.64.12, directory ‘/pub/GenomeDB/SubtiList’

A 21 548 bp nucleotide sequence around the 24° region of the Bacillus subtilis chromosome, located 306 kb downstream from the zero point of the physical map, was determined. Twenty-one putative ORFs were identified: two ORFs (Orf1 and Orf20) were consistent with the nucleotide and amino acid sequence of B. subtilis OrfJ, whose function is not known, and Pcp, respectively; four were found to display significant similarities to known proteins in data banks, i.e. 5-keto-3-deoxyglucarate dehydratase (Orf2), aldehyde dehydrogenase (Orf3) and two glucarate dehydratases (Orf4 and 5); three had considerable similarity to the sensor-regulator proteins of bacterial two-component signal transduction systems (Orf1, 11 and 12); two had considerably high levels of similarity to parts of known proteins (Orf13 and 19); and five showed low levels of similarity to known proteins in the data banks. The remaining six ORFs showed no similarity to known proteins.

The nucleotide sequence of approximately 10 kb at the 5' flanking region (32°) of srfAA of the Bacillus subtilis chromosome was determined. Eleven putative ORFs were identified. Three of them (orf6, orf7 and orf8) coincided with known B. subtilis genes (comJ, coml and tlpC ) encoding a competence-specific protein, a DNA-entry nuclease and a transducer-like protein, respectively. The products of two other ORFs showed similarity to GlnP of Escherichia coli (orf1) and β-glucosidase A of B. polymyxa (orf5).

The gene encoding extracellular xylanase (xynA) was amplified as a 770 bp DNA fragment from Bacillus subtilis 168 chromosomal DNA by PCR. The genes encoding endo-β-1,4-glucanase (eglS) and endo-β-1,3-1,4-glucanase (bglS) were isolated from a genomic library of B. subtilis 168. The sequences of xynA and eglS were identical to those of the xylanase and cellulase genes from B. subtilis PAP115. Integrative plasmids containing DNA fragments with deletions in the coding region of the genes were constructed and used to replace the chromosomal eglS, bglS and xynA genes of B. subtilis 168. Strains without any detectable activity against xylan (Xyn−), carboxymethylcellulose (Egl−) or mixed linked β-1,3-1,4-glucan (Egl− Bgl−) were obtained. The genes were mapped at 170° (eglS), 175° (xynA) and 340° (bglS) on theB. subtilischromosome.

SUMMARY: The gene encoding thymidylate synthase A (thyA) was cloned from a genomic library of Bacillus subtilis 168. The sequence of thyA was found to be highly similar to that of the phage ϕ3T thymidylate synthase gene. This similarity is, however, limited to about 862 nucleotides, spanning the coding region and the adjacent 3’ region. The flanking sequences are not related. An integrative plasmid containing a DNA fragment with a deletion within the coding region of thyA was constructed and used to replace the chromosomal thyA gene. Transformants unable to grow without thymidine at 47 °C were obtained. Genetic and physical mapping techniques were used to show that the cloned DNA fragment harbouring ΔthyA was integrated at 168° on the B. subtilis chromosome.

SUMMARY: The nucleotide sequence of 20 kb contiguous to the pksX locus of Bacillus subtilis was determined. Six ORFs were recognized, one of which extended for 13 341 nucleotides. Their predicted products have significant similarities to proteins with known functions involved in the synthesis of polypeptides and polyketides or in fatty acid metabolism. At the nucleotide level, three regions with a high level of sequence identity (49-54%) to the Aspergillus nidulans wA gene, responsible for the synthesis of a polyketide pigment, were recognized. The observed similarities suggest that the 20 kb region and the previously reported 13-6 kb region containing pksX are part of the same locus, possibly involved in secondary metabolism.

SUMMARY: A gene was found in Bacillus subtilis which encodes a protein highly homologous to the Escherichia coli rpsA gene product, the S1 ribosomal protein. The B. subtilis protein contains the domain responsible for binding to ribosomes and two S1 motifs, instead of four as found in the E. coli protein. The B. subtilis protein is similar in this way to the equivalent protein of plant chloroplast ribosomes, supposed to be the counterpart of E. coli S1. The gene is expressed during vegetative growth in B. subtilis at the transcriptional and translational levels, as judged by Northern hybridization and expression in a translational fusion with a reporter gene. In contrast to the E. coli situation, it can be inactivated without dramatic effects on cell viability. Southern hybridization of the B. subtilis DNA fragment encoding this gene revealed specific homologous fragments in all other Gram-positive bacteria tested. The hybridization pattern with B. stearothermophilus suggests the presence of at least two homologous genes in this bacterium. We show that in B. subtilis the ORF preceding the rpsA homologue encodes a protein which is highly similar to the product of the E. coli mssA gene which is located upstream of rpsA. Again, in contrast to the E. coli situation, where these genes are co-transcribed, in B. subtilis they are separated by a transcription terminator and the mssA homologue is transcribed during sporulation. We suggest that during the evolution very similar structures and genetic organization of these two genes were conserved but acquired different functions in Gram-negative and Gram-positive bacteria.

The dnaD gene of Bacillus subtilis was identified within a 104 kb DNA segment cloned into a yeast artificial chromosome. The nucleotide sequence of the wild type and dnaD23 mutant genes were determined. dnaD is predicted to encode a protein of 232 amino acids with no similarity to proteins in the data banks.

As part of the Bacillus subtilis genome sequencing project, we have determined the complete nucleotide sequence of a skin element which is located between spoIVCB and spoIIIC. The entire sequence of this element is 48032 bp in length, and contains 57 ORFs with putative ribosome-binding sites. Two of them correspond to previously sequenced and characterized genes, cwIA and spoIVCA. Furthermore, seven ORF products identified in this element show interesting similarities with known proteins present in data banks, including the ø105 immunity repressor, the ø105 Cro-like protein and the SPP1 terminase. These results indicate the possibility that the skin element is a cryptic remnant of an ancestral temperate phage.

The 29·71 kb chromosomal region of Bacillus subtilis 168 extending from 308° to 311° contains 18 ORFs. Functions of most of these ORFs were identified and associated with cell wall metabolism. Sequences of two non-coding regions of 0·7 and 2·2 kb flanking the ggaAB operon involved in the synthesis of poly(3-O-β-D-glucopyranosyl N-acetylgalactosamine 1-phosphate), a minor teichoic acid, correspond to five degenerate segments of neighbouring protein-coding regions. We discuss the possibility that such grey holes are indicative of a chromosomal rearrangement which could have arisen from horizontal gene transfer.

SUMMARY: Within the framework of an international project for the sequencing of the entire Bacillus subtilis genome, a 29 kb chromosome segment, which contains the hut operon (335°) and the wapA gene, has been cloned and sequenced. This region (28954 bp) contains 21 complete ORFs and one partial one. The 5th, 6th and 17th genes correspond to hutH encoding histidase, hutP encoding the positive regulator for the hut operon and wapA encoding a precursor of three major wall-associated proteins, respectively. A homology search for their products deduced from the 21 complete ORFs revealed that nine of them exhibit significant homology to known proteins such as urocanase (Pseudomonas putida), a protein involved in clavulanic acid biosynthesis (Streptomyces griseus), amino acid permeases (lysine, Escherichia coli; histidine, Saccharomyces cerevisiae; and others), β-glucoside-specific phosphotransferases (E. coli and Erwinia chrysanthemi) and 6-phospho-β-glucosidases (E. coli and Erw. chrysanthemi). Based on the features of the determined sequence and the results of the homology search, as well as on genetic data and sequence of the hut genes reported by other groups, it is predicted that the B. subtilis hut operon may consist of the following six genes (6th-1st), the last of which is followed by a typical ρ-independent transcription terminator: hutP, hutH, EE57A (hutU) encoding urocanase, EE57B (hutl) encoding imidazolone-5-propionate hydrolase, EE57C (hutG) encoding formiminoglutamate hydrolase and EE57D (tentatively designated as hutM) possibly encoding histidine permease. Interestingly, the direction of transcription of these hut genes is opposite to that of the movement of the replication fork.

SUMMARY: Increased productivity in DNA sequencing would not be valid without a straightforward detection and estimation of errors in finished sequences. The sequence of the surfactin operon from Bacillus subtilis was obtained by two different groups and by chance we were also working on the same chromosome region. Taking advantage of this situation we report in this paper, the number and nature of errors found in the overlapping part of the DNA sequences obtained by the three laboratories. The coincidence of some of the errors with compression in sequence ladders and with secondary DNA structures as well as the detection of frameshift errors using computer programs, are demonstrated. Finally we discuss the definition of a new sequencing strategy that might minimize both the error rate and the cost of sequencing.